An imaging sensor simulation model is described which allows a modeled or measured scene radiance map to be displayed on a video monitor as it would be seen if viewed through a simulated sensor under simulated environmental conditions. The model includes atmospheric effects (transmittance, path radiance, and single-scattered solar radiance) by incorporating a modified version of the LOWTRAN 6 code. Obscuration and scattered radiance introduced into the scene by battlefield induced contaminants are represented by a battlefield effects module. This module treats smoke clouds as a series of Gaussian puffs whose transport and diffusion are modeled in a semi-random fashion to simulate atmospheric turbulence. The imaging sensor is modeled by rigorous application of appropriate optical transfer functions with appropriate insertion of random system noise. The simulation includes atmospheric turbulence transfer functions according to the method of Fried. Of particular use to sensor designers, the various effects may be applied individually or in sequence to observe which effects are responsible for image distortion. Sensor parameters may be modified interactively, or recalled from a sensor library. The range of the sensor from a measured scene may be varied in the simulation, and background and target radiance maps may be combined into a single image. The computer model itself is written in FORTRAN IV so that it may be transported between a wide variety of computer installations. Currently, versions of the model are running on a VAX 11/750 and an Amdahl 5860. The model is menu driven allowing for convenient operation. The model has been designed to output processed images to a COMTAL image processing system for observer interpretation. Preliminary validation of the simulation using unbiased observer interpretation of minimum resolvable temperature (MRT)-type bar patterns is presented.
A procedure and system have been developed to interactively correct the geometry of image data, and merge the data with auxiliary graphical data. This system provides the following capabilities: 1. Correct the absolute and relative geometry of images. 2. Register two or more images to each other. 3. Register and overlay graphics data onto image data. 4. Change the geometry of the image and graphics data into any of twenty standard cartographic projections. 5. Provide an assessment of the accuracy of the geometric operation. The system provides for both manual selection of ground control points to establish the geometric correction parameters, and an automatic determination of mapping parameters from operator definition of the desired cartographic projection. Experiments have been conducted using a number of data sources, including Landsat Thematic Mapper data, geophysical gravity data, and digital line graph cartographic data.
Two classes of satelliteborne EO sensors have been extensively studied: staring sensors, and sensors that scan (slowly) by rotation of the entire satellite. Linear scanners are a less-studied but important intermediate type, featuring 3-axis stabilization of all or most of the satellite, relatively rapid scanning via a scan mirror, and TDI (time-delayed integration) for high sensitivity. Advantages include efficient use of solar arrays and heat-radiators, easy cross-link to other satellites, and much easier producibility than a staring sensor of comparable resolution. An existing sensor-simulation program was modified to simulate linear scanners (in addition to its previous simulation options), and allow free choice of key parameters such as scan period, scan orientation, threshold, NET, optical blur, etc. Other parameters are derived from these in a consistent way. Existing programs for simulating the processing of sensor data were modified to process the outputs of the linearscan simulator. End-to-end simulations were successfully performed for targets, noise, and synthetic background (derived from real scenes). The assumed form of onboard signal processing is a bandpass analog filter per TDI channel, followed by sampling, A/D conversion, and peak-detection. Alternates will be developed as future options.
The design of multispectral scanners optimized for a certain job or set of jobs is being done more frequently now that multispectral collection and exploitation technologies are maturing. In this paper, the conceptual design procedure is reviewed. Then the role of computer simulation is illustrated using the example of the design study for the Thematic Mapper. Although this study was performed ten years ago, it still illustrates the principles of the use of computer simulation.
A set of electro-optical sensor simulation models is integrated into an interactive, userfriendly software package designed for use in modeling and simulation of electro-optical sensor systems and mosaic focal plane arrays. The software system, Improved Optical Detection System (IODS), includes modules for image and signal generation, image degradation due to optical system and environmental MTF characteristics, detector spectral, temporal and spatial frequency response, signal modulators, and processing electronics transfer function characteristics. The software system is currently utilized for beam diagnostics, surveillance, and remote sensing applications.
Laser imagers, operating at long ranges, have recently become of interest for both military and civilian applications. Systems using large optics and powerful sources are expensive to build and test, and so the image signal to noise and resolution performance should be investigated analytically before testing. However, in many cases, a simulation may provide a better idea of the actual performance. The simulation described here is for a monostatic laser-illuminator angle-angle imager, which is essentially a flash camera. Important phenomena which are modeled include surface reflectivity and glints, diffraction, photon noise, detector noise, and detector size. Laser speckle and illumination beam profiles have not been incorporated into this model. The images resulting from a simulation can additionally answer the question, "What do we do with the image after we get it?", which cannot be addressed by signal to noise or resolution calculations. The model was developed and can be used on an IBM PC, without any special purpose hard-ware designed specifically for image processing.
A study was done to evaluate the pointing performance of Space Telescope with a quadrant detector used as the fine guidance sensor. The detector, a quadrant digicon, has been proposed as a replacement detector should unforseen problems develop with the present baseline interferometer design. The detector model is discussed along with the experimental data from which it was derived.
Use of computer modeling to predict performance and production yield of photodiode arrays has been previously reported. In this earlier work, a narrow, collimated slit source of radiation was assumed to be normally incident on the back side of a generalized photodiode array. An analysis of carrier diffusion in the array led to values for diode quantum efficiency and line spread function, which in turn led to values for absolute signal responsivity, noise equivalent irradiance, and modulation transfer function. A further modeling extension is reported here which brings the model a step closer to real life electro-optical systems. The former idealized radiation input to the focal plane is replaced by a radiation blur circle generated by a ray trace model. The ray trace model simulates radiation spatial, angular, and spectral characteristics of an operational frontend optics module. A methodology for interfacing ray-trace models to carrier diffusion models is considered here, with attention to some radiation transport details which can significantly influence model predictions.
With the ever advancing trends towards more highly complex proposed data sensing systems coupled with satellite onboard distributed computing systems, there is a need for sophisticated, yet simple-to-use methods and tools for modeling the workload capacities of the data processing components of these systems well before actual implementation. These data processing components will receive data from different types of space-based sensor systems such as starers, spinners, linear scanners, and step starers. In order to aid systems engineers in the design of these advanced systems, a number of commercially available program modeling tools were reviewed, but for our specific purpose, one modeling tool was found to be most effective for modeling and analyzing different onboard processing architecture configurations/concepts for future space-based satellite surveillance systems. This tool, for our purpose, was found to provide flexibility in designing space-based surveillance systems that promote survivability, endurability, and data integrity and to identify and eliminate many potential system performance problems before implementation.
Computer aided design as an engineering tool has permeated all aspects of the infrared sensor industry. The availability and low cost of computer design tools allows a wide spectrum of engineering challenges to be impacted by computer aided design. This paper presents a practical view of the power and applicability of computer aided design. This paper draws on current design experience for infrared detector performance modeling, analog multiplexer performance analysis, integrated circuit mask layout, and infrared sensor test. Practical examples of each design task are reviewed.
Large-aperture lightweight space optics must be designed from first principles so as to compensate for a multitude of disturbing influences if they are to perform to approximate theoretical perfection during a long period of life. Active control of structures and active optical components such as deformable mirrors are vital when dynamic performance is an issue. This overview deals with a hypothetical (but not unreal) nominal system, handling disturbances, and adaptive optics. The 2.4-meter Hubble Space Telescope primary mirror is held up as an example of an active deformable mirror designed with the latest mathematical tools.
Asymmetric aspheric optical surfaces are very difficult to fabricate using classical techniques and laps the same size as the workpiece. Opticians can produce such surfaces by grinding and polishing, using small laps with orbital tool motion. However, hand correction is a time consuming process unsuitable for large optical elements. Itek has developed Computer Controlled Optical Surfacing (CCOS) for fabricating such aspheric optics. Automated equipment moves a nonrotating orbiting tool slowly over the workpiece surface. The process corrects low frequency surface errors by figuring. The velocity of the tool assembly over the workpiece surface is purposely varied. Since the amount of material removal is proportional to the polishing or grinding time, accurate control over material removal is achieved. The removal of middle and high frequency surface errors is accomplished by pad smoothing. For a soft pad material, the pad will compress to fit the workpiece surface producing greater pressure and more removal at the surface high areas. A harder pad will ride on only the high regions resulting in removal only for those locations.
Rockwell has realized rapid testing of Infrared Focal Plane Arrays (IRFPAs) using a totally automated cryogenic test station with the latest technology in device handling, data acquisition, illumination and throughput capabilities. This station provides testing of HgCdTe Focal Plane Arrays fabricated in a fully certified production facility. All aspects of this facility are under Quality Control surveillance including the hardware and software used by the automated test station.
Rockwell International Corporation has established an automated database system for the storage, retrieval and analysis of focal plane manufacturing information. In 1983, Rockwell implemented the first fully certified Mercury Cadmium Telluride (HgCdTe) production facility in the United States. As part of the development of this facility, key processing variables were identified for inclusion in a comprehensive data base. Our database system contains information from boule, slice, wafer and die sources. The current database consists of data from over one million on-line measurments drawn from the entire range of our production and test facilities. Database analysis routines provide various tabular, plot and statistical output that are used toward the goals of lowering cost and improving yield in the process of HgCdTe manufacturing.
This paper is concerned with the development of Lens Design Techniques, specifically related to the use of personal computers. Lens Designers have been using computers for more than 20 years, although for most of this time it has only been feasible to use large mini-computers and mainframes, usually in a time-sharing mode. Since about 1980 there have been an increasing number of personal computers available, and these are now used very widely for Lens Design, as well as for many other applications.
The Technical Computer Center has been providing computer aided engineering support from the origin of Orlando Aerospace engineering. These computer aids have evolved as the computer technologies have evolved. Historically, TCC's hybrid lab has supported Systems Engineering activities of trajectory analysis, autopilot and guidance analysis, and design verification. Also, TCC's MADRE lab has traditionally supported engineering data acquisition, reduction, and analysis. More recently the capabilities of TCC have been expanded to meet the requirements of Electro-optic, Electronic, and Mechanical engineering analysis and design.